simistic assumptions about life in the state of nature, government's superiority .... Hoppe, DemocracyâThe God That Failed (New Brunswick, N.J.: Trans- action ...
Journal of Libertarian Studies Volume 17, no. 4 (Fall 2003), pp. 17–37 2004 Ludwig von Mises Institute www.mises.org
DO PESSIMISTIC ASSUMPTIONS ABOUT HUMAN BEHAVIOR JUSTIFY GOVERNMENT? Benjamin Powell* and Christopher Coyne** Is government a necessary institution?1 To answer this question, many theorists begin with an account of a state of nature in which there are no “rules” or institutions to regulate human behavior. They then compare that state of nature with outcomes that are theoretically achieved with a monopoly rule enforcer—a government.2 Assumptions vary widely about how humans will behave in the absence of a rule-enforcing institution. Some utopian accounts claim that everybody will cooperate and never have violent interactions.3 In pessimistic accounts of the state of nature, people have no morality or respect for life or property, and will cheat, kill, and steal at every opportunity.4 One account even assumes that humans would behave *
Assistant Professor of Economics, San Jose State University. Research Fellow at the James M. Buchanan Center, George Mason University, and Social Change Fellow at the Mercatus Center. 1 Unlike many authors we consider herein, we do not equate “anarchy” with the absence of institutions of governance. Anarchy is defined as the absence of a coercive monopoly government. The state of nature is a subset of this condition, where there are no voluntary, private enforcing institutions or laws. 2 See, e.g., Thomas Hobbes, Man and Citizen (Indianapolis, Ind.: Hackett, 1991), p. 115; and James Buchanan, The Limits of Liberty (Indianapolis, Ind.: Liberty Fund, 2000), pp. xiv, 8. 3 For an example, see Pierre-Joseph Proudhon, What is Property?, ed. and trans. Donald R. Kelley and Bonnie G. Smith (Cambridge: Cambridge University Press, 1994) 4 Hobbes, Man and Citizen, pp. 115, 118. **
17
Journal of Libertarian Studies as the children do in the fictional book Lord of the Flies or like rats that are deprived of food in a laboratory experiment.5 In this article, we do not attempt to adjudicate between competing hypotheses about human behavior in the state of nature. We assume the worst about human behavior, and see if, even with the most pessimistic assumptions about life in the state of nature, government’s superiority can be theoretically established. Consistency and methodological individualism dictate, though, that what we assume about individual morality and motives in the state of nature, we must also assume about the individuals we analyze running governments. We begin by examining a pessimistic account of the state of nature. Hobbes’s description, in Leviathan and in Man and Citizen, is reflected in the economic descriptions of the state of nature used by James Buchanan, and also by Martin McGuire and Mancur Olson, when they justify the foundation of government. Next, we look critically at the social contract solution that Buchanan proposes to see if it is feasible in such a world.6 We then consider whether McGuire and Olson’s account of a stationary bandit is necessarily superior to even Hobbes’s pessimistic vision of the state of nature.
A PESSIMISTIC ACCOUNT OF THE STATE OF NATURE Although Hobbes did not believe in the natural rights associated with libertarianism today, he nonetheless rejected the idea that people acted immorally in the state of nature. He thought they acted within their rights, but that their rights hopelessly conflicted. Hobbes wrote: Nature hath given to every one a right to all; that is, it was lawful for every man, in the bare state of nature, or before 5
See Thomas Hogarty, “Cases in Anarchy,” in Explorations in the Theory of Anarchy, ed. Gordon Tullock (Blacksburg, Va.: Center for the Study of Public Choice, 1972). For a response to Hogarty’s article, see Virgil Storr, “Defining Anarchy as Rock ’N Roll,” in Explorations in the Public Choice Theory of Government, ed. Edward Stringham (forthcoming). 6 For different, but not inconsistent, critiques of Buchanan’s social contract justifications for government, see Walter Block and Thomas DiLorenzo, “Is Voluntary Government Possible? A Critique of Constitutional Economics,” Journal of Institutional and Theoretical Economics 156, no. 4 (2000), pp. 567–82; and Walter Block and Thomas DiLorenzo, “Constitutional Economics and The Calculus of Consent,” Journal of Libertarian Studies 15, no. 3 (2001), pp. 37–56.
18
Powell & Coyne – Pessimistic Assumptions such time as men had engaged themselves by any covenants or bonds, to do what he would, and against whom he thought fit, and to possess, use, and enjoy all what he would, or could get.7
It is only natural that men, in exercising their rights in a world of scarcity, will come into conflict. Hobbes recognizes this, writing: The most frequent reason why men desire to hurt each other, ariseth hence, that many men at the same time have an appetite for the same thing; which yet very often they can neither enjoy in common, nor yet divide it; whence it follows that the strongest must have it, and who is strongest must be decided by the sword.8
Hobbes’s state of nature will naturally lead to a situation where, he says, “it cannot be denied but that the natural state of men, before they entered into society, was a mere war, and that not simply, but a war of all men against all men.”9 From this perpetual war in the state of nature, we arrive at the famous Hobbesian Jungle, where life is nasty, brutish, and short. This Hobbesian framework is Buchanan’s analytical starting point in The Limits of Liberty. Buchanan’s own ethical and normative political concern, unlike Hobbes’s, is respect for the autonomy of the individual person.10 For Buchanan, anarchy is ideal, but the tendency toward violation of others, absent an enforcer, makes it impractical. The individualist must view any reduction in the sphere of activities ordered by anarchy as an unmitigated “bad.” He must recognize, nonetheless, that anarchy remains tolerable only to the extent that it does produce an acceptable degree of order. The anarchistic war of each against all, where life becomes nasty, brutish, and short, will be dominated by the order that the sovereign can impose.11
In examining the state of nature, Buchanan looks solely at a person’s economic incentives, acknowledging, but excluding, any moral reasons that might influence someone not to steal or commit an act of
7
Hobbes, Man and Citizen, p. 117. Hobbes, Man and Citizen, p. 115. 9 Hobbes, Man and Citizen, p. 118. 10 Buchanan, The Limits of Liberty, p. xiv. 11 Buchanan, The Limits of Liberty, p. 8. 8
19
Journal of Libertarian Studies aggression. He frames the individuals’ choices in a prisoners’ dilemma game, where the numbers in the cells reflect A’s and B’s utility. Figure 1. Person B Respects Property
Person B Doesn’t Respect Property
Person A Respects Property
19, 7
3, 11
Person A Doesn’t Respect Property
22, 1
9, 2
In Figure 1, all people are better off if everyone engages only in productive activities, but it is individually optimal to plunder others. The Nash equilibrium is the Hobbesian result, where both people are worse off than if they had respected each other’s property. In a simple two-person interaction, Buchanan recognized that the people could make a self-enforcing agreement to respect each others rights: “Each person may respect the agreed-on assignment because he predicts that defection on his part will generate parallel behavior by the other party.”12 Due to repeated plays, each person will abide by the contract, knowing that if they do not, the other will also defect. The problem society faces is one with many anonymous individuals wherein the optimal strategy is to defect from the agreement and steal from others. Buchanan wrote: As more parties are added to the initial contractual agreement, in which an assignment of rights is settled, the influence of any one person’s behavior on that of others becomes less and less. . . . In large-number groups, each individual rationally acts as if his own behavior does not influence the behavior of others. He treats others’ behavior as part of his natural environment, and he adjusts his behavior accordingly. . . . Each person has a rational incentive to default; hence, many persons can be predicted to default and the whole agreement becomes void unless the conditions of individual choice are somehow modified.13
12
Buchanan, The Limits of Liberty, p. 85. Buchanan, The Limits of Liberty, p. 85.
13
20
Powell & Coyne – Pessimistic Assumptions Since it is individually optimal to defect, the society plunges back into the Hobbesian jungle.14 Because of the individual incentive for defection, Buchanan concludes that a third-party enforcement mechanism is necessary to assure that individuals will respect property rights. He believes that everybody would conceptually agree to a social contract that forms a state in order to get out of the Hobbesian Jungle.15 Olson’s and McGuire’s roving bandit model also derives from a Hobbesian framework.16 The underlying assumption of this analysis is that under anarchy, roving bandits will destroy the incentives for people to invest and produce. Olson and McGuire claim that social wealth is maximized if the bandits become stationary bandits by establishing themselves as dictators over specific geographic areas.17 14 However, the transition from small-number settings to large-number settings may not be as clear as Buchanan makes it, because large groups are made up of many interconnected personal bilateral exchanges. The amount of overall defection, even in large groups, may be limited by the reluctance to defect in each of these two-person exchanges. 15 He does not consider the possibility of private non-monopoly third-party enforcement, but instead moves directly to the necessity of the state. Although Buchanan seems to think that the only type of enforcement mechanism must be a monopoly government, there is plenty of evidence in the literature of other private voluntary enforcement mechanisms. For historical accounts, see Bruce Benson, “Enforcement of Private Property Rights in Primitive Societies: Law Without Government,” Journal of Libertarian Studies 9, no. 1 (1989), pp. 1–26; David Friedman, “Private Creation and Enforcement of Law—A Historical Case,” Journal of Legal Studies (March 1979), pp. 399–415; and Terry Anderson and P.J. Hill, “America’s Experiment With Anarcho-Capitalism: The NOT so Wild, Wild West,” Journal of Libertarian Studies 3, no. 1 (1979), pp. 9–29. Theoretical descriptions can be found in Murray N. Rothbard, For a New Liberty (San Francisco: Fox & Wilkes, 1973); David Friedman, The Machinery of Freedom (La Salle, Ill.: Open Court, 1989); Edward Stringham, “Market Chosen Law,” Journal of Libertarian Studies 14, no. 1 (1999), pp. 53–77; and Hans-Hermann Hoppe, “The Private Production of Defense,” Journal of Libertarian Studies 14 (Summer 1998), pp. 27–54. 16 Mancur Olson, “Dictatorship, Democracy, and Development,” American Political Science Review 87 (1993), pp. 567–76; Mancur Olson, Power and Prosperity (New York: Basic Books, 2000); and Martin McGuire and Mancur Olson, “The Economics of Autocracy and Majority Rule: The Invisible Hand and the Use of Force,” Journal of Economic Literature 34 (1996), pp. 72–96. 17 Although not the main focus of this paper, one must also question the rationality of the move from the roving band to the stationary band. As Hoppe
21
Journal of Libertarian Studies They conclude that stationary bandits will be guided by an invisible hand to provide stability and public goods. It is as if the ruling power were guided by a hidden hand no less paradoxical for us than the invisible hand in the market was for the people in Adam Smith’s time. . . . The invisible hand will lead it, remarkably, to treat those subject to its power as well as it treats itself.18
According to the authors, a central government has a direct interest in the product of the ruled. It has the incentive to provide a stable social order including protection as well as other public goods. Taxes are kept to a level at which the deadweight loss is smaller than the gains in government revenue. The result is increased welfare for all parties involved. Since the only alternative available they consider is the initial state of nature in which roving bandits destroy all incentives to produce and create a general state of chaos, McGuire and Olson apparently believe that government creates the best achievable state of affairs.
CAN A SOCIAL CONTRACT WORK IN A HOBBESIAN WORLD? Life in the state of nature with no third-party enforcement and immoral individuals would be bleak. Individuals would benefit if they could constrain each other from plundering. Buchanan’s solution to the problem is for the people to “conceptually” agree to a social contract establishing a monopoly third-party enforcer. Buchanan writes: demonstrates, when individuals are able to protect, defend, and insure themselves against crime (i.e., the roving bandit), the effect on time preference is “temporary and unsystematic.” However, when government (i.e., a stationary bandit) becomes involved, the effect on time preferences as compared to crime is systematically different and more profound. Given that government expropriation is constant while crime is intermittent, with a monopoly government, property right violations become institutionalized and the rate of return on future investments is adjusted downward. See Hans-Hermann Hoppe, Democracy—The God That Failed (New Brunswick, N.J.: Transaction, 2001), pp. 12–15. 18 McGuire and Olson, “The Economics of Autocracy and Majority Rule,” pp. 73–74. If this claim is taken to its logical conclusion, it literally means that that there exists no distinction between being a receiver of taxes and a payer of taxes, for if there is a distinction, then no equal treatment exists. (We are indebted to Hans Hoppe for this point.)
22
Powell & Coyne – Pessimistic Assumptions The design and location of this [enforcement] institution becomes all important, however; neither party will entrust enforcement to the other, and, indeed, the delegation of such authority to one party in contract violates the meaning of enforcement.19
The optimal third party is some type of automated machine outside of the game. Absent the possibility of an impartial machine, an outside referee is optimal. Buchanan likens this to a baseball umpire: In the game analogy that we have used several times before, the protective state is the umpire or referee, and, as such, its task is conceptually limited to enforcing agreed-on rules.20
Although this may be ideal, how well does the analogy of a baseball umpire apply to government as an interpreter and enforcer of rules? The nature of the social contract, as described by Buchanan, is that everyone in a particular area unanimously agrees to the assignment of rights, and then forms a government to enforce the assigned rights.21 The government, however, is not like the baseball umpire. The game being played is life, and the players, by the nature of the social contract, are all the people in a given location, including those who work for the government; the umpires are players themselves. This is the exact situation in which, according to Buchanan, people would never put themselves: entrusting enforcement to another party to the contract. The government, as a third party, has the job of interpreting and enforcing the rules in discrepancies between players, but the government is also the only interpreter and enforcer that determines what duties the social contract gave the government itself. Buchanan is well known for applying economic analysis to government operations, and for modeling political actors as self-interested individuals who respond to incentives.22 However, in The Limits of
19
Buchanan, The Limits of Liberty, pp. 120–21. Buchanan, The Limits of Liberty, p. 206. 21 In some writings, Buchanan moves away from actual unanimity and writes of “relative unanimity” or “80 percent unanimity.” See James Buchanan and Gordon Tullock, The Calculus of Consent (Ann Arbor: University of Michigan Press, 1962), p. 107. Rothbard points out this misleading use of semantics. See Murray N. Rothbard, “Buchanan and Tullock’s The Calculus of Consent,” in Applications and Criticism from the Austrian School, vol. 2 of The Logic of Action (Cheltenham, U.K.: Edward Elgar, 1997), p. 271. 22 See, e.g., Buchanan and Tullock, The Calculus of Consent. 20
23
Journal of Libertarian Studies Liberty, he does not analyze government with the same assumptions that he makes about people in anarchy. When Buchanan characterizes government as “conceptually external,” he no longer analyzes it as if it were comprised of individuals with their own purposes. Once we recognize that the government must be comprised of people, we are forced to apply public-choice-style analysis to this government. Since particular assumptions about individuals behaving in their own selfinterest, with no regard for morality, were made in analyzing the state of nature, we must also assume these same things about the people who will run the government. Similarly, Robert Nozick recognizes this basic point when he suggests that we might want to use the “minimax” criteria in comparing anarchy to states: The state would be compared with the most pessimistically described Hobbesian state of nature. But in using the minimax criterion, this Hobbesian situation should be compared with the most pessimistically described possible state, including future ones. Such a comparison, surely, the worst state of nature would win.23
In any social contract, individuals obviously would not choose to form an all-powerful leviathan government that could exploit them. Instead, they would want to form a government limited to enforcing their agreement not to steal from each other. This begs an important question: Will the government adhere to the initial social contract? If the government interprets and enforces the contract, there is no third-party enforcement constraining it. Because of this, Alfred Cuzan maintains, we never get out of anarchy. Instead, we shift from market anarchy to political anarchy, thereby actually increasing the amount of violence in society: There is no “third party” to make and enforce judgments among the individual members who make up the third party itself. The rulers still remain in a state of anarchy vis-à-vis each other.24
In the social contract, the government is in a position similar to what was faced in the original prisoner’s dilemma in Figure 1, with
23
Robert Nozick, Anarchy, State, and Utopia (New York: Basic Books, 1974), p. 5. 24 Alfred Cuzan, “Do We Ever Really Get Out of Anarchy?” Journal of Libertarian Studies 3, no. 2 (1979), p. 152.
24
Powell & Coyne – Pessimistic Assumptions one crucial difference. If the government defects, the other citizens do not have the same option to defect because they still face an enforcer: the government. The game now can be modeled as in figure 2.25 Figure 2. Citizen Obeys Social Contract
Citizen Doesn’t Obey Social Contract
Government Obeys Social Contract
19, 7
18, 1
Government Doesn’t Obey Social Contract
25, 1
24, 0
Following the assumptions from the anarchy model, people, including government officials, only act in line with the economic incentives they face. Even though the initial social contract or “constitution” may have only provided for a limited government, the people in government will defect on that contract because they are the enforcers. There is no third party constraining them.26 A representative citizen is modeled on the top in Figure 2. If the government abides by the agreed-on contract, the citizen will achieve a utility level of “7.”27 Instead of another citizen, the government is now modeled on the side.28 The government may depart from the contract, although it will 25
It has been suggested that figure 2 should be a two-box diagram because a citizen would not be able to defect on a contract if the government were effective. But even against an effective government, citizens can always defect from a contract, although they will be punished for doing so. The government modeled in this diagram is effective in the sense that there is a 100% probability of punishment if the citizen does defect. That is why the payoffs force the citizen to choose to obey the social contract. 26 Even if a “veil of ignorance” (or uncertainty) is assumed, the result does not change. The citizens could write an optimal social contract (constitution) not knowing what their positions will be. As soon as the contract is put into effect, some particular people must fill the roles of government to enforce the contract. They will now know who they are and will defect on the contract that was initially written behind a veil of ignorance. 27 We do not mean to imply cardinal utility. Ordinal rankings are sufficient to achieve the represented outcomes. 28 There is still the two-citizen game of the type that Buchanan models. With government enforcing contracts, both citizens will abide by their contracts
25
Journal of Libertarian Studies interpret it to be within the contract, and confiscate the citizen’s wealth. If the citizen attempts to defect after seeing that the government has defected, the government will use its enforcement monopoly, find him guilty of deviating from the contract, and take away his remaining utility. (The government does not take all the utility when it first departs from the contract, lest it leave the citizen with no incentive to continue abiding by the contract.) The above example shows only one citizen’s choice set against a government that departs from the social contract. The threat of many citizens simultaneously departing from the contract, in a rebellion, might provide some form of enforcement on the government. However, the individual calculus that each citizen faces when deciding to participate makes this a weak constraint because a public-goods problem has to be overcome.29 In the real world, morality and ideology help to overcome the public-goods problem, but Buchanan assumes these things away so only the narrow economic calculus can be examined.30 Therefore, the threat of rebellion will not force government to abide by the social contract. In the simple prisoner’s dilemma game above, government was modeled as a single individual. However, in reality, governments are comprised of many individuals with a separation of duties and powers among them. Randy Barnett characterized the checks-and-balances system as follows: The essence of this strategy is to create an oligopoly or a “shared” monopoly of power. This scheme preserves a monopoly of power but purports to divide this power among a number of groups.31
Nonetheless, the above model can still accurately represent government. Even with a separation of powers between different branches of government, or between different levels of government through federalism, there are gains to be had from cooperation among the branches with each other. However, Buchanan never models the second prisoner’s dilemma of “social” contract between the government and the citizen. 29 See Gordon Tullock, “The Paradox of Revolution,” Public Choice 11 (1971), pp. 89–100. 30 For an account of the role of ideology in private national defense that would apply to rebellion, see Jeffery R. Hummel, “The Will to Be Free: The Role of Ideology in National Defense,” Independent Review 5, no. 4 (2001), pp. 523–37. 31 Randy Barnett, The Structure of Liberty (New York: Oxford University Press, 1998), p. 253.
26
Powell & Coyne – Pessimistic Assumptions or levels. Even though, in an individual round of play, the interests of different government branches may be in opposition to each other, over multiple plays they can all gain in utility from cooperating to expand the power of government. Rothbard described the checks-and-balances system by writing: As we have discovered in the past century, no constitution can interpret or enforce itself; it must be interpreted by men. And if the ultimate power to interpret a constitution is given to the government’s own Supreme Court, then the inevitable tendency is for the Court to continue to place its imprimatur on ever-broader powers for its own government. Furthermore, the highly touted “checks and balances” and “separation of powers” in the American government are flimsy indeed, since in the final analysis all of these divisions are part of the same government and are governed by the same set of rulers.32
With no one to enforce the social contract on the government, it will not remain limited. It will transform into the leviathan government that Hobbes envisioned. Buchanan himself seems to recognize this at one point: There is no obvious and effective means through which the enforcing institution or agent can itself be constrained in its own behavior. Hence, as Hobbes so perceptively noted more than three centuries ago, individuals who contract for the services of enforcing institutions necessarily surrender their own independence.33
But Buchanan then moves away from this position and continues in the remainder of the book to model government as conceptually external. The state emerges as the enforcing agency or institution conceptually external to the contracting parties and charged with the single responsibility of enforcing agreed-on rights and claims along with contracts which involve voluntarily negotiated exchanges of such claims.34
Buchanan’s limited social contract is not an option if the world is populated by Hobbesian individuals, because government is not external.35 32
Rothbard, For a New Liberty, p. 48. Buchanan, The Limits of Liberty, p. 87. 34 Buchanan, The Limits of Liberty, p. 88. 35 It is only because of Buchanan’s break from methodological individualism that he is able to reach his conclusions. Rothbard, reviewing The Calculus of 33
27
Journal of Libertarian Studies It is an active participant and can be expected to defect from any social contract made.36 It remains possible that this leviathan government, or autocracy, may nevertheless be preferable to the state of nature. The state will have the ability, because of its monopoly over the use of force, to seize more than anybody could in the state of nature.37 It may refrain from maximum short-term predation in order to extract more in the long run by leaving the people more wealth to maintain their incentive to produce. Buchanan’s model leaves us with a choice between many roving bandits or one large stationary bandit who is able to steal more than any roving bandit, but who may limit his short-term theft in order to extract more in the long term. This is, in fact, Olson’s model of government and the state of nature. With Hobbesian assumptions, limited government is not an option. Consent and discussing the use of “social costs,” wrote that “despite much talk by Buchanan and Tullock of their staunch individualism, especially methodological individualism, they are not consistent individualists at all.” Rothbard, “Buchanan and Tullock’s The Calculus of Consent,” p. 273. 36 This is not to say that no enforcement mechanism is possible in a Hobbesian world. The key is that there must be a competing force capable of constraining it. Much work has been done on polycentric systems of enforcement since Buchanan’s original work. Work should be done to establish how well these systems could function in a world populated by Hobbesian individuals. For detailed analysis of many of the workings of polycentric systems, see Rothbard, For a New Liberty; Friedman, Machinery of Freedom; Bruce Benson, To Serve and Protect (New York: New York University Press, 1998); Bruce Benson, Enterprise of Law (San Francisco: Pacific Research Institute for Public Policy, 1990); Stringham “Market Chosen Law”; Barnett, The Structure of Liberty, pp. 238–97; and Hoppe, “The Private Production of Defense.” For recent debates about power relations, networks, and collusion in a polycentric system, see Tyler Cowen, “Law as a Public Good: The Economics of Anarchy,” Economics and Philosophy 8 (1992), pp. 249–67; David Friedman, “Law as a Private Good: A Response to Tyler Cowen on the Economics of Anarchy,” Economics and Philosophy 10 (1994), pp. 319–27; Daniel Sutter, “Asymmetric Power Relations and Cooperation in Anarchy,” Southern Economic Journal 61 (1995), pp. 602–13; Tyler Cowen and Daniel Sutter, “The Costs of Cooperation,” Review of Austrian Economics 12 (1999), pp. 161–73; and Bryan Caplan and Edward Stringham, “Networks, Law, and the Paradox of Cooperation,” Review of Austrian Economics 16, no. 4 (2004). 37 “Historically, by far the overwhelming portion of all enslavement and murder in the history of the world have come from the hands of government.” Rothbard, For a New Liberty, p. 47.
28
Powell & Coyne – Pessimistic Assumptions We next consider whether a leviathan government is necessarily superior to the state of nature.
DO THE INTERESTS OF THE RULERS AND THE RULED DOVETAIL? The key idea in McGuire’s and Olson’s analyses is that the ruler will act as an income-maximizing, rational economic agent. We agree that the ruler may attempt to maximize his monetary income, and in so doing will provide stability and keep taxes at a level that minimizes the deadweight costs. In many cases, though, the ruler will maximize not only monetary revenue but also psychic income.38 Psychic income includes non-pecuniary forms of income that are valued as well, as Rothbard describes: Each individual acts so that the expected psychic revenue, or achievement of utility, from his action will exceed its psychic cost. The latter is the forgone utility of the next best alternative that he could adopt with the available means. Both the psychic revenue and psychic cost are purely subjective to the individual.39
The nature of psychic income is such that outside observers are unable to assign, a priori, specific characteristics that constitute psychic income for the ruler. The implications of this realization are that rulers may gain psychic income from things other than money, such as the very act of holding and wielding power itself.40
38
For another account of how time preferences affect ruler behavior but that does not go to the extremes of McGuire and Olson, finding in fact that a decentralized system of natural elites is superior to monopoly government, see Hoppe, Democracy—The God that Failed. The analysis herein is not inconsistent with his analysis that the interests of monarchs or others with a long-term hold on the monopoly of power may more closely dovetail with their citizens’ interests then will the interests of leaders with only a short-term hold on the monopoly of power. Our analysis is simply leading us away from the conclusion of Olson and McGuire that, by establishing a monopoly enforcer, the interests of the leader are close enough to his citizens to necessarily make it superior to a situation with no monopoly of power. 39 Murray N. Rothbard, Man, Economy, and State (Auburn, Ala.: Ludwig von Mises Institute, 1993), p. 273. 40 For a general model of dictatorship that models rulers gaining utility from both their own consumption and wielding power, see Ronald Wintrobe, The
29
Journal of Libertarian Studies Psychic income creates a potential disconnect between the interests of the rulers and those of the ruled. A ruler may repress the ruled even though he may have low-time preferences and a stable hold on his position because he gains psychic income by doing so. Of course, psychic income can affect ruler behavior in both directions. It is possible that the ruler could gain psychic income from doing “good” and bolstering his self-image as a paternal protector. We do not wish to contend this point; rather, by incorporating psychic income into the analysis, we see that the rulers are not necessarily guided by an invisible hand to provide for the ruled. It is possible that they may do so, but it is also possible that they may exploit the ruled. In either case, the potential to cause great harm cuts against the strong conclusion of the parallel of ruler and ruled interests that McGuire and Olson wish to tell.41 The claim that McGuire and Olson can make is severely limited once psychic income is included in the analysis. They assert that the ruling group will be guided by an “invisible hand” to treat the ruled group in a better-than-expected manner—in some cases, as well as it treats itself. But we can no longer make that claim once we realize that the rulers will not always act to maximize monetary income alone. The authors’ contention that the rulers will act in such a manner is predicated on the assumption that they are monetary-income maximizers. If we accept that the rulers may act to maximize not just monetary income, then conclusions stemming from this assumption are highly questionable as well. This realization serves to explain the repressive behavior of many dictators observed in the real world.42 Wintrobe sums up this point:
Political Economy of Dictatorship (Cambridge: Cambridge University Press 1998), chap. 5. 41 The notion of psychic income applies to the ruled as well as to the ruler. One cannot assume that the ruled aim solely to maximize monetary income, for they, too, maximize non-pecuniary forms of income. The realization that the notion of psychic income applies to both parties further highlights the disconnect between their interests. It is by no means clear that they are as aligned as McGuire and Olson assume. 42 For a discussion of the many examples, both past and present, of governments acting as predators on the populace, see Douglass North, “Institutions and Credible Commitment,” Journal of Institutional and Theoretical Economics 149 (1993), pp. 11–23.
30
Powell & Coyne – Pessimistic Assumptions The assumption of revenue maximization can result in an underestimate of the damage autocrats can do as well as a misleading idea of the benefits of stationarity if applied to some dictatorships. To put it simply and to take only the most obvious cases, we can say that the Jews under the Nazis, the blacks in South Africa, and the peasants under Stalin would all undoubtedly, were they given the choice, have preferred that their bandits were a little less stationary.43
Realizing that rulers maximize psychic income is also consistent with Glaeser’s and Shleifer’s work finding that elected officials in a redistributive democracy can, and in many cases have, supported policies of “destructive taxation” and public-good provision aimed at maximizing their chances of reelection.44 Government officials blatantly neglect the interests of the ruled in order to maintain their position of power.45 Provision of public goods is a key element of the stationary bandit since Olson and McGuire assume that government-provided public goods are needed for the generation of any and all income.46 A public good or collective good is characterized by nonrival consumption (indivisibility) and nonexcludability, where there is no marginal cost for an additional person consuming the good. It is a good in that it provides a positive benefit. However, identifying the existence of public goods becomes problematic once psychic income is considered, because once psychic income is incorporated into the analysis, the good is not necessarily positive and indivisible. Surely there will be at least one individual who is against the provision of the good by a coercive government, and this person will suffer psychic harm by its mere provision. Such an individual does not receive a positive collective good or service, but rather suffers an individual harm. Since the possibility exists that some individuals are hurt by the coercive provision of the good, then it cannot really be a collective service characterized as being indivisible and positive. For the good or 43
Wintrobe, The Political Economy of Dictatorship, p. 132. Edward Glaeser and Andrei Shleifer, “The Curley Effect,” NBER Working Paper 8942 (2002). 45 Hoppe, in Democracy—The God that Failed, when analyzing various government regimes, finds this divergence of interests will be even greater in a democracy. 46 McGuire and Olson, “The Economics of Autocracy and Majority Rule,” p. 74. 44
31
Journal of Libertarian Studies service that brings harm is indeed both separable (divisible) and negative to the individuals that suffer.47 The additional unit is not costless, but rather imposes a psychic cost upon the suffering individual.48 While recognizing the divergence of interests between the ruler and the ruled, it may be argued that if one considers a long-run equilibrium, the psychic income of rulers is not a relevant issue. The parallelism of the interests of the ruled and the rulers would come about, over time, through the elimination of “bad” rulers. McGuire and Olson envision a competitive process where the ruled move between jurisdictions to more congenial political systems. They incorporate perfect-competition assumptions into their model claiming, ceteris paribus, that the interests of the ruled and rulers dovetail in the long run. However, mobility between jurisdictions is not enough to ensure that rulers incorporate the preferences of the ruled into decisions about taxation and public-good provision. When local governments raise revenues through the property tax, they avoid Tiebout-style competitive pressures from other local governments because the tax is simply capitalized into real estate values.49 Citizen mobility does not allow people to escape excessive taxation; they either stay and pay the tax directly, or sell out and pay the tax via a lower sale price. Even with perfect mobility, once a tax is imposed, it becomes a sunk cost that owners either pay directly or through the lower sale price they receive on their property, and then the new owner transfers money to the government.50 The property tax allows competitive 47
See Murray N. Rothbard, “The Myth of Neutral Taxation,” in Applications and Criticism from the Austrian School, pp. 56–108. 48 Putting the issue of psychic income aside, there is still the problem of finding the optimal level of public good provision. Like all other goods, public goods are provided in marginal units, not in a homogeneous lump. When the government providing the good must determine the level of production of the good, it suffers from the economic calculation problem that is absent on the market. In the market, producers and entrepreneurs are guided by the profitand-loss mechanism to serve customer wants in the most efficient manner. This guiding mechanism is absent for the government, as is any judicious criterion for determining the level of good provision. See Ludwig von Mises, Bureaucracy (Grove City, Penn.: Libertarian Press, 1983), p. 53. 49 Bryan Caplan, “Standing Tiebout on his Head: Tax Capitalization and the Monopoly Power of Local Governments,” Public Choice 108 (2001), pp. 101–22. 50 If a citizen has expectations of future tax increases that others do not know about, he will have an incentive to move so he can avoid taxation. If others
32
Powell & Coyne – Pessimistic Assumptions local governments to transform into monopolists because, while they may be “competing” to get mobile citizens, the land in their jurisdiction is fixed and is a source of wealth that can be appropriated. Similarly, when placed on property owners, other forms of taxation can also avoid competitive pressures by capitalizing into land values.51 Even with unrealistic assumptions of perfect information and no transaction costs, competition between jurisdictions will not force rulers’ interests to align with the interests of the ruled. Leaders will be able to extract wealth from land values against citizens’ wishes. McGuire and Olson assume that the ruled have perfect information about the rulers, not only in their current jurisdiction but in other jurisdictions as well. In a dictatorship that oppresses its citizens, however, a self-interested ruler would attempt to limit the amount of information citizens have about the standard of living in other jurisdictions. Also, since the other “competing” jurisdictions are stationary bandits and citizens are often too afraid to express their displeasure with their own ruler, even the information that citizens in one jurisdiction are able to get about another jurisdiction is likely to be unreliable. The assumption of perfect information limits the ability of the model to help explain the workings of any realistic political and economic order. We must also question why a stationary bandit would allow the ruled to move to other jurisdictions as they wished. With a monopoly on the use of force, a stationary bandit could easily create barriers to exit, leaving the ruled little choice even if they do not like the policies of their ruler and would prefer to move to another jurisdiction. A stationary bandit with preferences other than the maximization of monetary income can impose great harm on the ruled. Tax capitalization, imperfect citizen information, and the ability to limit citizen migration all cause the interests of stationary bandits and their citizens to diverge. Since the ruler has a monopoly on the use of force, he can inflict much greater harm than any individual in the state of nature. Consider, for example, that the harm inflicted by totalitarian regimes
also have access to the information, then these expectations will be capitalized into real estate values. In either case, the land, being a fixed source of wealth that the government is going to tax, remains in the jurisdiction, so the government can extract rents from it against citizens’ wishes. 51 On this point, see Benjamin Powell, “Tax Capitalization and the Monopoly Power of Local Governments: An Extension of Caplan (2001),” Public Choice (forthcoming).
33
Journal of Libertarian Studies in the twentieth century caused approximately 169 million deaths.52 When power is monopolized in the hands of a few individuals, rulers have the ability to cause much greater harm than when power is dispersed. Even given the pessimistic assumptions regarding the state of nature, it is not clear that a coercive government improves the situation.
CONCLUSION We have considered two accounts of the evolution from a state of nature to a social order that allows for interaction and exchange. Both accounts begin with pessimistic assumptions about individuals in the absence of a central government. In both cases, the authors conclude that the state, with a monopoly on force, is in the interest of all in society. We have demonstrated that Buchanan’s argument and McGuire’s and Olson’s argument, even under pessimistic assumptions, fail to show that government is necessarily superior to an unfavorable state of nature. They fail to realize that their theoretical accounts do not allow them to escape the problems that stem from their own pessimistic assumptions. The existence of a central state shifts and magnifies the power structure that is present in the initial state of nature. Rather than power being dispersed among the populace, it is centralized in the hands of a few. Once individuals possess this power, it is far from clear that they will promote the interests of the ruled. In some cases they may, but in others they may not. And, in the cases where they do not, they have the power to impose great harm on others. Given this uncertainty, one is unable to conclude, contrary to Buchanan and to Olson and McGuire, that a central state is better than an institutionless state of nature. The Hobbesian vision of anarchy has been extremely influential. Our contribution here has been to show that two attempts by publicchoice theorists to use this Hobbesian vision to prove government’s necessity are fatally flawed. This does not, in and of itself, prove the superiority of anarchy. The question of whether human behavior in an institutionless anarchy is really as bad as Hobbes predicts is a historical question for others to answer. There is also the possibility for private competitive third-party enforcement to improve on the Hobbesian result even with pessimistic assumptions. We are optimistic about both of these possibilities. Our point is merely that even with the worst
52
R.J. Rummel, Death by Government (New Brunswick, N.J.: Transaction, 1994), Table 1.2.
34
Powell & Coyne – Pessimistic Assumptions assumptions about human nature, government, on purely theoretical grounds, cannot be shown necessarily to improve the situation.
BIBLIOGRAPHY Anderson, Terry L., and P.J. Hill. “American Experiment in AnarchoCapitalism: The Not So Wild, Wild West.” Journal of Libertarian Studies 3, no. 1 (1979). Barnett, Randy. The Structure of Liberty. New York: Oxford University Press, 1998. Benson, Bruce. “Enforcement of Private Property Rights in Primitive Societies: Law Without Government.” Journal of Libertarian Studies 9, no. 1 (1989). ———. The Enterprise of Law: Justice without the State. San Francisco: Pacific Research Institute for Public Policy, 1990. ———. To Serve and Protect. New York: New York University Press, 1998. Block, Walter, and Thomas DiLorenzo. “Constitutional Economics and The Calculus of Consent.” Journal of Libertarian Studies 15, no. 3 (2001). ———. “Is Voluntary Government Possible? A Critique of Constitutional Economics.” Journal of Institutional and Theoretical Economics 156, no. 4 (2000). Buchanan, James. The Limits of Liberty. Indianapolis, Ind.: Liberty Fund, 1975. Buchanan, James, and Gordon Tullock. The Calculus of Consent. Ann Arbor: University of Michigan Press, 1962. Caplan, Bryan. “Standing Tiebout on His Head: Tax Capitalization and the Monopoly Power of Local Governments.” Public Choice 108 (2001). Caplan, Bryan, and Edward Stringham. “Networks, Law, and the Paradox of Cooperation.” Review of Austrian Economics 16, no. 4 (2004). Cowen, Tyler. “Law as a Public Good: The Economics of Anarchy.” Economics and Philosophy 8 (1992). Cowen, Tyler, and Daniel Sutter. “The Costs of Cooperation.” Review of Austrian Economics 12 (1999). Cuzan, Alfred. “Do We Ever Really Get Out of Anarachy?” Journal of Libertarian Studies 3, no. 2 (Summer 1979). Friedman, David. “Law as a Private Good: A Response to Tyler Cowen on the Economics of Anarchy.” Economics and Philosophy 10 (1994).
35
Journal of Libertarian Studies ———. The Machinery of Freedom. La Salle, Ill.: Open Court, 1989. ———. “Private Creation and Enforcement of Law: A Historical Case.” Journal of Legal Studies (March 1979). Glaeser, Edward L., and Andrei Shleifer. “The Curley Effect.” NBER Working Paper 8942 (2002). Hobbes, Thomas. Leviathan. Cambridge: Cambridge University Press, 1996. ———. Man and Citizen. Indianapolis, Ind.: Hackett, 1991. Hogarty, Thomas. “Cases in Anarchy.” In Explorations in the Theory of Anarchy, edited by Gordon Tullock. Blacksburg Va.: Center for Study of Public Choice, 1972. Hoppe, Hans-Hermann. Democracy—The God That Failed. New Brunswick, N.J.: Transaction, 2001. ———. “The Private Production of Defense.” Journal of Libertarian Studies 14 (Summer 1998). Hummel, Jeffery. “The Will to Be Free: The Role of Ideology in National Defense.” Independent Review 5, no. 4 (2001). McGuire, Martin, and Mancur Olson. “The Economics of Autocracy and Majority Rule: The Invisible Hand and the Use of Force.” Journal of Economic Literature 34 (1996). Mises, Ludwig von. Bureaucracy. Grove City, Penn.: Libertarian Press, 1983. North, Douglass. “Institutions and Credible Commitment.” Journal of Institutional and Theoretical Economics 149 (1993). Nozick, Robert. Anarchy, State, and Utopia. New York: Basic Books, 1974. Olson, Mancur. “Dictatorship, Democracy, and Development.” American Political Science Review 87 (1993). ———. The Logic of Collective Action: Public Goods and the Theory of Groups. Cambridge, Mass.: Harvard University Press, 1971. ———. Power and Prosperity. New York: Basic Books, 2000. Powell, Benjamin. “Tax Capitalization and the Monopoly Power of Local Governments: An Extension of Caplan (2001).” Public Choice. Forthcoming. Proudhon, Pierre-Joseph. What is Property? Edited and translated by Donald R. Kelley and Bonnie G. Smith. Cambridge: Cambridge University Press, 1994. Rothbard, Murray N. “Buchanan and Tullock’s The Calculus of Consent.” In Applications and Criticism from the Austrian School. Vol. 2 of The Logic of Action. Cheltenham, U.K.: Edward Elgar, 1997.
36
Powell & Coyne – Pessimistic Assumptions ———. For a New Liberty. San Francisco: Fox & Wilkes, 1973. ———. Man, Economy, and State. Auburn, Ala.: Ludwig von Mises Institute, 1993. ———. “The Myth of Neutral Taxation.” In Applications and Criticism from the Austrian School. Vol. 2 of The Logic of Action. Cheltenham, U.K.: Edward Elgar, 1997. Rummel, R.J. Death by Government. New Brunswick, N.J.: Transaction, 1994. Storr, Virgil. “Defining Anarchy as Rock ’N Roll: Rethinking Hogarty’s Three Cases.” In Explorations in the Public Choice Theory of Government, edited by Edward Stringham. Forthcoming. Stringham, Edward. “Market Chosen Law.” Journal of Libertarian Studies 14, no. 1 (1999). Sutter, Daniel. “Asymmetric Power Relations and Cooperation in Anarchy.” Southern Economic Journal 61 (1995). Tullock, Gordon. “The Paradox of Revolution.” Public Choice 11 (1971). Wintrobe, Ronald. The Political Economy of Dictatorship. Cambridge: Cambridge University Press, 1998.
37
